Question: I would like to ask you one question regarding your differentiation between “symptomatic” drugs like monoamine-agonists or dopamine-inhibitors and second messenger modulators, the latter showing some ability to modify the disease process as such. From my clinical point of view I can very much confirm the fact, that lithium and valproate are one of the best drugs we have in the long-term run.
My question would be: Why is that actually? I can find no huge difference in the way both classes of the drugs work in terms of mode of action. So called antidepressants work via GPCRs, modulating downstream effects and in the long term alter cell functions like neurogenesis, neuroprotection and so forth. As do second messenger modulators. Although i did some extensive research on downstream modulation I think it boils down to the same alteration of cell function in both cases. I just don’t get the difference you are implying.
More concretely: What would be the net effect for a neuron to be modulated by changes in GSK3ß, Bcl-2 etc. or CREB/BDNF etc.? Maybe I am missing some basic insights here (if yes please put me out of my misery by providing some literature) but I think it doesn’t add up to justify a disease-modifying effect of 2nd messenger modulators just by the modulation of 2nd messengers since other psychoactive drugs do that as well. What do you think?
PL: Thank you for this profound question. The reply I’ll give you here is not definitive, but merely my thoughts currently. It is worthy of scientific papers and discussion in the scientific literature, but unfortunately this topic has not received sustained attention in psychiatry yet.
So my current views are that the concept of disease modification (DM) can be defined in two ways: clinically and biologically. The clinical definition is, in my view, more solid scientifically, although the limited discussion on disease modification in psychiatry so far is influenced by an FDA discussion, which tended to focus more on the biological part, as you do.
Let’s first discuss the clinical concept and then I’ll return to the biological aspect which is central to your question.
Clinically, DM means altering the course of the illness, and especially improving morbidity and mortality, not just improving symptoms. This is understood clearly in cardiology, especially since there the issue of even improving symptoms is moot. Cardiology is not about improving the symptoms of chest pain or dyspnea during heart attacks. Rather, it’s about preventing heart attacks and reducing mortality. Lipid-lowering drugs and antihypertensives have those disease-modifying effects. But they do not have symptomatic benefits. Nitroglycerin has symptomatic pain-reducing benefit during a heart attack, but not usually long-term disease modification benefits. We know the pathophysiology of heart disease involves cholesterol buildup in arteries and narrowing of arteries; hence the effects of statins and antihypertensives also supports their disease modifying effects. However, I would add, even if we did not know why or how statins or antihypertensives reduced heart attacks and cardiac mortality, those clinical benefits by themselves would be sufficient to consider those drugs to be disease-modifying.
In psychiatry, we know that lithium reduces frequency of mood episodes, and reduces mortality overall, especially by suicide. These clinical effects are analogous to lipid lowering drugs as above, and thus would be disease-modifying clinically. We know that monoamine blockers and dopamine blockers do not have those clinical episode prevention benefits nor do they reduce suicide or mortality. So it is clear clinically that lithium has disease modification and monoamine agonists and dopamine blockers do not. That could be the end of the story, because the clinical reality is what matters most, even if we don’t know the biological mechanisms of such clinical effects. Nonetheless, most people, including the FDA, are invested in the notion of biological mechanisms.
Turning to biology, the question would be how does lithium affect the known disease process in bipolar illness? And are those effects part of the pathophysiology of bipolar illness? This question is more complex, and the answer is not as definitive as in cardiology, but it is not completely unknown either. Goodwin and Jamison’s large second edition textbook on Manic-Depressive Illness has extensive sections running into hundreds of pages on the genetics and pathophysiology of bipolar and unipolar mood illness. We know a lot about the pathogenesis of these conditions. For instance, we know that circadian rhythms are disturbed strongly in bipolar illness. Lithium directly affects the CLOCK gene, which influences circadian rhythms, and it normalizes CLOCK gene function in an animal model of mania.
Regarding second messengers, lithium differs from most other drugs, especially monoamine agonists and dopamine blockers, because its effects are direct, not indirect. Lithium is a direct second messenger modifier, and it extensively affects many different second messengers. Monoamine agonists and dopamine blockers affect second messengers only indirectly, via their initial effects on first messengers, the neurotransmitters and their effects on postsynaptic receptors, which are then translated by G protein second messengers into the cell. In contrast, lithium (and valproate) directly stimulate or inhibit various G proteins, and also go directly into the axon to stimulate or inhibit a host of other second messengers. These effects are direct and much more extensive than with dopamine blockers and monoamine agonists, from a cellular perspective.
How does this translate into disease modification? Do second messenger effects directly lead to mood episode prevention? We don’t know really, but they could. For instance, the second messenger effects translate to long-term changes in neuronal transcription which changes axonal connections and literally the microanatomy of the brain in those regions where lithium has effects. These long-term changes could be the basis for lithium’s long-term clinical effects. But we cannot say so for certain.
About two decades ago, Frederick Goodwin and I linked the available literature on second messengers to the concept of long-term course benefits. We didn’t know think of the concept of disease modification versus symptomatic effects at that time in an explicit way, but we had a similar idea in mind. (https://pubmed.ncbi.nlm.nih.gov/9951560/)